Understanding Transient Left Ventricular Ejection Fraction Reduction During Atrial Fibrillation With Artificial...

Area of Science:

• Cardiology
• Artificial Intelligence in Medicine
• Echocardiography

Background:

• Atrial fibrillation (AF) can cause a temporary reduction in left ventricular ejection fraction (LVEF) that typically resolves with the restoration of sinus rhythm.
• Beat-to-beat variability in AF can lead to mismeasurement of LVEF, complicating accurate assessment.
• The prognostic significance of true transient LVEF reduction during AF requires further investigation.

Purpose of the Study:

• To determine the frequency of mismeasured LVEF in patients with AF due to beat-to-beat variability.
• To assess whether genuine transient LVEF reduction during AF has prognostic implications for heart failure hospitalization.

Main Methods:

• Observational study analyzing patients with echocardiograms in AF and subsequent sinus rhythm within 90 days.
• Classification of patients based on LVEF changes: no reduction, transient reduction, or persistent reduction.
• Utilized an AI algorithm for automated multicycle LVEF measurement to reclassify patients and Fine-Gray hazard modeling for 1-year heart failure hospitalization risk.

Main Results:

• Of 810 patients, 34.6% had persistent LVEF reduction, 8.8% had transient reduction, and 56.7% had no reduction.
• AI reassessment increased AF-LVEF by 8.2%, reclassifying 28.2% of patients.
• Transient AF-LVEF reduction, as identified by AI, was associated with a significantly higher 1-year heart failure hospitalization risk (HR 2.28).

Conclusions:

• Artificial intelligence may reduce misdiagnosis of reduced LVEF during AF.
• AI can more accurately identify true transient AF-LVEF reduction.
• Transient AF-LVEF reduction represents a potentially high-risk phenotype requiring further clinical attention.